Performance evaluation of polygon-based and voxel-based finite element analysis on tibia with osteogenesis imperfecta

Abstract

Osteogenesis Imperfecta (OI) is a bone disease that commonly can cause bone deformation and could cause death in extreme cases. To treat this disease, rodding surgeries are required. Such surgeries require pre-surgery planning to evaluate the fracture risk. Finite element analysis (FEA) has been widely used as computational tool to predict bone fracture. It is known that the accuracy of FEA is strongly dependent on discretization of element. There are two common types of element that is being used for bone analysis, namely polygon-based and voxel-based. Current clinical practices are more inclined to creating voxel-based FE models based on patient specific computed tomography (CT) image as it is more accessible to them. Polygon-based FE models are less commonly used as medical personnel usually does not has the engineering expertise in handling FEA. However, there might be benefits in employing polygon-based models. This paper aims to compare the FEA performance between both types of FE models, and then suggest the better option for clinical usage. It was found that in most cases, the FEA predictions of both models were similar with percentage difference of <1%. However, polygon-based models showed its advantage when it comes to computing demand at it was able to perform FEA as much as 45 times faster than voxel-based model. To conclude, polygon-based FE model could be the better option for clinical uses for offering identical accuracy with less computing demand.